SOLVED Crucial MX300 SSD SMART failure

[Klontje]

The TBW should be correct by the way according to the FreeNAS statistics. See image below, 379T written. I guess I'll have to figure out which jail is responsible for this (first thing to pop into my mind, Zoneminder) and move the working folder to HDD storage on my regular pool.

 

[Johnnie Black]

The TBW should be correct by the way according to the FreeNAS statistics. See image below, 379T written.

Yep, 379TiB = 416.7TB, that's a lot of write activity and it will kill any SSD quickly.

 

[Klontje]

Replaced one of the two with a fresh one that just arrived. Data shows indeed a counter for RAW_VALUE going up and for VALUE going down. I will keep the old one running as long as I can next to the fresh one and keep the other new one on stock to prevent them from running out at the same time again in the future. Also found the culprit, have Zoneminder running in a jail processing multiple HD video streams. Data is stored for analysis purposes before it is written off to its final destionation which is already on my regular HDD based pool. I guess I'm gonna have to move that scratch space to HDD storage as well. Thanks for thinking this through with me and I hope this is a good thread for anybody looking as to why their SSD's are wearing out as fast as they are.

 

[Licretia3]

Another clue that the ssd is near its end of life is its very high Average Block Erase Count (one of the SMART attributes). On my Crucial MX500 ssd, each increase of 15 results in a 1% decrease of Remaining Life.

I reduced the frequent writing of temporary files to ssd by moving them to a hard disk. In some cases, the corresponding app is hard-coded to expect its files to be located on the system drive (the C: ssd in my Windows system) but setting up symlinks solved that problem. An example is the CyberPower UPS software, which frequently writes UPS status to a folder on C: and needed a symlink. I'm not a user of FreeNAS but googling tells me it supports symlinks too.